Second the idea of introducing some layout algorithms On Tuesday, 21 February 2012 22:27:24 UTC+8, Michael Hunger wrote: > > It has been discussed to parallelize traversals but so far not much effort > has gone into that. > > I think right now the most probably candidate for parallelizing traversals > is cypher as its declarative structure > allows to reason about the necessary number of parallel traversals and how > to combine their results. We would probably > look into Scala's parallel collection libraries to achieve that. > > GPU's are interesting for parallelizing but you have to shuffle the graph > data into the GPU first, which might need some effort. > > Another interesting GPU related topic is vizualization, using the GPU > (CUDA) to calculate the graph-layout and then opengl again to actually > render it. > Would love to see something happening there . > > Cheers > > Michael > > Am 21.02.2012 um 15:21 schrieb Ashwin Raghav: > > > Hey, > > I wanted to find out whether there are any projects/plans around > > parallelizing graph traversal. I did learn from one of the InfoQ > > lectures by Emil that there was effort being put into parallelizing > > search but did not find any information beyond that. > > > > I'd be happy to know what steps are being taken around this. Has there > > been any work done around parallelizing search using GPUs. > > > > Ashwin > > > > Twitter: @ashwinraghav > > blog.ashwinraghav.com > >
-- You received this message because you are subscribed to the Google Groups "Neo4j" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
